DESCRIPTION (provided by the PI): Comparative effectiveness data that quantify effects of central nervous system stimulants (stimulants) on real-life function in the treatment of attention deficit/hyperactivity (ADHD) disorder are lacking. A relevant clinical outcome greatly affected by ADHD is driving performance, specifically crashes and traffic citations, but studies have been restricted to evaluations under testing conditions and have not considered actual driving records. In addition, safety concerns about fatal and near-fatal cardiac side effects are currently based on limited evidence, leaving appropriate clinical and regulatory action heavily debated. Our group recently completed a retrospective study of a 10-year Florida Medicaid cohort, which found an increased risk for emergency department visits for cardiac symptoms, but sample size limited investigations of severe events or the impact of coexisting cardiac risk factors. We propose to expand our previous research with a retrospective cohort study of more than 500,000 youth with ADHD eligible for Medicaid benefits in 14 large states (representing 16 million youth and 70% of the entire pediatric Medicaid population) between 1999 - 2006 to: (1) evaluate the cardiac risk of stimulants overall, and of mixed amphetamine salts (MAS) versus methylphenidate (MPH) individually, on fatal or severe cardiovascular events;(2) assess whether stimulant risk is modified when used concomitantly with antipsychotics, antidepressants, or clonidine;(3) evaluate stimulant effectiveness overall, and between MAS versus MPH individually, on reduced risk for traffic citations and crashes. Data on medication use, relevant risk factors and clinical endpoints will be ascertained from Medicaid Extract Files, the National Death Index, and Division of Motor Vehicle driving records. A new-user design will be used that compares periods of stimulant use against periods of non-use and former use. Several sensitivity analyses will explore potential effects of misclassification and residual (unadjusted) confounding. Results will contribute evidence to the comparative risk and benefit of stimulants with special consideration of vulnerable pediatric populations, such as those with complex psychiatric needs as well as minorities, and help guide clinical policies and practice and inform future research concerning the community management of ADHD.